Despite the emergence of multiple and extensively drug resistant strains, tuberculosis (TB) is largely a curable disease if treated appropriately. However, in some tuberculosis-endemic areas, fewer than 40% of TB cases are diagnosed due to the lack of accurate and easy-to-use diagnostic assays, and these continue to drive the epidemic. Currently, diagnosis relies on demonstration of the bacteria, Mycobacterium tuberculosis, in clinical specimens by sputum smear microscopy and repeat culture. Limitations of culture often include slow growth (up to 8 weeks), low sensitivity as a result of required decontamination procedures, difficulty in obtaining serial samples in children and non-sputum producers, and the high cost of the assay. A non-sputum, non-growth based biomarker assay could replace microbiologic intermediate endpoints, could transform the pace and scope of TB drug development and of global TB control. It may additionally have utility for reaching a population of paucibacillary disease as is often seen in children, extrapulmonary TB, and HIV/TB. This proposed research will establish our goal towards developing a highly sensitive approach to the quantitative detection of lipoarabinomannan (LAM), a cell envelope lipoglycan in serum and urine with clinical diagnostic accuracy. The project aims are 1) to optimize the urine and serum pretreatment protocol by testing proteases and chaotropic agents, thus making maximal amount of the LAM available for detection; 2) to generate highly specific and unique monoclonal antibodies directed against ?in vivo? forms of LAM; and 3) to take forward non-sputum based LAM detection via a capture ELISA with further validation in well-characterized adult cohorts with suspected TB in countries which have high burdens of TB and HIV (South Africa, Thailand, Kenya) and (Peru) with low HIV exposure. Given the recognized imperfection of culture as a gold standard, in a separate exploratory analysis we will determine ELISA sensitivity and specificity using LAM detection by Gas Chromatography/Mass Spectrometry (GC/MS) as the reference standard. Our capture ELISA accuracy targets are ?85% sensitivity and ?95% specificity. In addition, we will determine, the changes in urine and serum LAM concentrations during TB treatment; this will use a separate set of existing specimens and is intended to establish whether LAM in urine might be useful as a biomarker of treatment response. Our ultimate goal is to develop an accurate, simple Point-of-Care test based on quantitative engineering of capture chemistry and conditions for LAM detection in non-sputum specimens. The test should reach all TB cases irrespective of their HIV status. The chosen demonstration will not be a final device but will demonstrate a proof-of-principle that is aligned with the design goals.